High-precision grating ruler displacement demodulation algorithm and sensitivity calibration method

Abstract

The invention provides a high-precision grating ruler displacement demodulation algorithm and a sensitivity calibration method, which are characterized in that a system is composed of a grating ruler module (1), a sensitivity calibration method (2) and a high-precision displacement demodulation algorithm (3), and the grating ruler module (1) comprises a collimation light source (11), a scale grating (12), an indication grating (13) and a photoelectric detection device (14); the sensitivity calibration method (2) comprises an adjustable driving power supply (21), a uniform light source (22), a power meter (23) and a photoelectric detection device (24). Wherein the high-precision displacement algorithm (3) comprises a displacement demodulation algorithm (31) and a sensitivity demodulation algorithm (32). A sensitivity demodulation algorithm (32) demodulates the sensitivity of each pixel according to the voltage and optical power of each pixel of a photoelectric detection device (24) output by a sensitivity calibration method (2), a displacement demodulation algorithm (31) corrects the voltage of each pixel of the photoelectric detection device (14) output by a grating ruler module (1) by calling the sensitivity of each pixel, and calculates the light intensity E and the displacement x, wherein the displacement x corrects a nonlinear error caused by sensitivity difference. The method can be used for measuring high-precision infinitesimal displacement, and can be widely applied to the fields of precision semiconductor manufacturing, precision displacement measurement, precision motion control, closed-loop feedback control of numerical control machine tools and the like.

Description

(I) Technical field

[0001] The present invention relates to a high-precision grating ruler displacement demodulation algorithm and sensitivity calibration method, which can be used for high-precision micro-displacement measurement such as precision semiconductor manufacturing, precision displacement measurement, precision motion control, and closed-loop feedback control of CNC machine tools, and belongs to the field of photoelectric detection technology. (II) Background technology

[0002] With the advent of industrial modernization, people have put forward requirements for displacement measurement technology such as large range and high resolution. Laser interferometer and grating ruler are one of the ideal measurement methods that meet the above requirements. Laser interferometer uses a stable laser wavelength as the measurement reference. When the external environment such as temperature and air pressure changes, the refractive index of the medium will change, resulting in measurement errors and low stability, which is not conducive to applications in precision measurement and other fields. On the contrary, the grating ruler uses a stable grating pitch as the measurement reference and has higher robustness to changes in the external environment. The system has a more compact structure, is easy to install and debug, and has a lower manufacturing cost, making it widely used in fields such as CNC machining technology. With the rapid development of CNC machining technology, the technical level of grating rulers has been greatly promoted.

[0003] Grating rulers use the grating pitch as the measurement reference. Depending on whether there is an obvious diffraction effect on the light field passing through the grating, they can be divided into grating rulers based on geometric moiré fringes and grating rulers based on light diffraction. Taking the grating ruler based on moiré fringes as an example, when the scale grating as the measurement reference overlaps with the indicator grating on the reading head with relative displacement, the collimated light will form a periodic spatial light field after being modulated by the two gratings (moiré fringes, see Figure 2 ). According to the geometric shading principle, the fringe spacing A and direction Θ of the moiré fringe (such as Figure 2 As shown, θ is the angle relative to the vertical direction, with counterclockwise as positive and clockwise as negative. θ1 and θ2 in the formula satisfy the same rules) satisfy the following equation: Where a1, a2 and θ1, θ2 are the grating pitch and stripe direction of the two gratings respectively.

[0004] Formulas (1) and (2) represent the general pattern of moiré fringes (also called oblique moiré fringes). Depending on the values ​​of a1, a2 and θ1, θ2, the fringes will also present different patterns: when a1 = a2 = a and θ1 is close to θ2, the moiré fringes are almost perpendicular to the direction of the grid lines, and this type of fringes is called horizontal moiré fringes; when a1 is close to a2 and θ1 is close to θ2, the moiré fringes are parallel to the direction of the grid lines, and this type of fringes is called vertical moiré fringes. The horizontal moiré fringe spacing A ⊥ and longitudinal moiré fringe spacing A ∥ satisfy: In the above two styles, since the parameters on the denominator are small, the moiré fringe spacing A is much larger than the grating pitch a. When there is a small relative displacement of the grating pitch a between the scale grating and the indicator grating on the grating ruler, the moiré fringe will shift by a larger fringe spacing A, thus amplifying the displacement signal.

[0005] For more special optical gate moiré fringes (a1=a2 and θ1=θ2), according to formula (1), the moiré fringes spacing tends to infinity, and a multi-field scanning method is required to obtain the light field information of the incremental code channel. For multi-field scanning with a number of N (N≧3), N collimated light sources are required, and the incremental code channel of the indicator grating needs to open N windows aligned with the collimated light sources. The grating pitch and grating line direction of each window are the same as those of the scale grating. The grating lines of each window are sequentially 1 / N grating pitches apart to detect the light field signal with a phase difference of 2π / N. There is a group of photoelectric detection devices at the back end of each window to receive the light intensity signal of these N phases. Corresponding to multi-field scanning, the incremental window of the single-field scanning indicator grating only needs to open one window, and the back end of the window uses a densely arranged linear array of photoelectric detection devices, and the adjacent pixels receive light field signals with a fringe spacing difference of 1 / N. Since the incremental window of the single-field scanning only needs to open one window, the volume is small and easy to integrate. In addition, each phase of a single-field scan is detected by multiple pixels on the photoelectric detection device, which means that the average effect between pixels with the same phase can effectively reduce the error caused by environmental changes.

[0006] In an ideal situation, Figure 4 As shown in (a), the moiré fringes should be a spatial light field distribution in the form of a sine wave, and the light intensity on each phase pixel satisfies: Where φ = 2πx / a, which represents the phase shift associated with the incremental track displacement x of the grating scale. Adding E b To meet the condition E m > 0. The cosine part I and the sine part Q of the phase shift satisfy: Then, we perform square sum and inverse tangent operations: The light intensity E and phase shift φ can be obtained. For the common four-field scanning mode (ie, N = 4), formulas (6) and (7) degenerate into: The back-end circuit quantifies the voltage signal of each pixel on the photoelectric detection device, which is proportional to the light intensity, and substitutes it into the above formula for calculation to obtain the displacement x of the incremental code channel of the grating scale.

[0007] Whether it is multi-field scanning or single-field scanning, multiple sets of specially designed photoelectric detection devices (i.e., multiple pixels) are required to detect the light intensity at multiple different phases. Each pixel in the photoelectric detection device can be a photocell with a photovoltaic effect, or a module consisting of a photodiode loaded with a negative bias and a transimpedance amplifier, which can generate a voltage according to the irradiance of the light field irradiated thereon. However, due to the influence of the processing and manufacturing process of the photoelectric detection device, the sensitivity of each pixel (here refers to the proportional relationship between the output voltage and the input light field irradiance) is difficult to be consistent. At this time, the voltage value on each pixel becomes: where k m Represents the sensitivity of each phase pixel. The corresponding cosine part I and sine part Q change as follows: Since each sensitivity k m Different from the above formula, it is impossible to extract it outside the summation sign. Substituting the light intensity E and phase shift φ calculated by formulas (8) and (9) will introduce nonlinear errors. Figure 3 The blue dots in (a) and 3(b) indicate the phase shift φ calculated by substituting equations (13) and (14) into equation (9) and its error from the true value when there is a maximum 5% difference in sensitivity between the four phase pixels of the photodetector device under four-field scanning. The error can exceed 1% of the grating pitch.

[0008] In order to solve the above problems, the present invention discloses a high-precision grating scale displacement demodulation algorithm and sensitivity calibration method, which can be used for measuring high-precision tiny displacements and can be widely used in precision semiconductor manufacturing, precision displacement measurement, precision motion control, closed-loop feedback control of CNC machine tools and other fields. Figure 1The system block diagram of the present invention, wherein the photoelectric detection device (14) and the photoelectric detection device (24) are the same photoelectric detection device when they are respectively applied to the grating ruler module (1) and the sensitivity calibration method (2), and both can output a resolvable voltage signal according to the irradiance of the light field incident thereon. The difference is that the photoelectric detection device (14) receives a spatially modulated light field, and the photoelectric detection device (24) receives a spatially uniform light field. The sensitivity demodulation algorithm (32) in the high-precision displacement demodulation algorithm (3) can calculate the sensitivity (usually in V / W) of each pixel of the photoelectric detection device according to the voltage signal output by the photoelectric detection device (24) and the light power output by the power meter (23). The calculated sensitivity can be converted into a floating point number with a unit of 1 through appropriate normalization operations, and used as a relative value of the sensitivity for the displacement demodulation algorithm (31) to call. The displacement demodulation algorithm (31) first reversely calculates the light intensity E received by each pixel in the photoelectric detection device (14) based on the voltage signal output by the photoelectric detection device (14) and the sensitivity calculated by the sensitivity demodulation algorithm (32). m , and then by substituting formulas (6) and (7) into formulas (8) and (9), the corrected light intensity E and phase shift φ are demodulated, and the displacement x on the high-precision grating scale incremental code track is obtained. (III) Summary of the invention

[0009] The purpose of the present invention is to provide a high-precision grating scale displacement demodulation algorithm and sensitivity calibration method to solve the nonlinear error caused by the difference in sensitivity of different pixels on the photoelectric detection device, which can be used for high-precision micro-displacement measurement. Figure 1 As shown, the system consists of a grating ruler module (1), a sensitivity calibration method (2), and a high-precision displacement demodulation algorithm (3). The grating ruler module (1) includes a collimated light source (11), a scale grating (12), an indicator grating (13), and a photoelectric detection device (14); the sensitivity calibration method (2) includes an adjustable driving power supply (21), a uniform light source (22), a power meter (23), and a photoelectric detection device (24); and the high-precision displacement algorithm (3) includes a displacement demodulation algorithm (31) and a sensitivity demodulation algorithm (32).

[0010] The objective of the present invention is achieved in the following way: the photoelectric detection device (14) and the photoelectric detection device (24) are the same photoelectric detection device respectively applied to the grating ruler module (1) and the sensitivity calibration method (2), and are respectively used to receive the spatially modulated light field and the spatially uniform light field, and both can output a resolvable voltage signal according to the irradiance of the light field incident thereon; the sensitivity demodulation algorithm (32) in the high-precision displacement demodulation algorithm (3) can calculate the sensitivity (usually in V / W) of each pixel of the photoelectric detection device according to the voltage signal output by the photoelectric detection device (24) and the light power output by the power meter (23); the calculated sensitivity can be converted into a floating point number with a unit of 1 through appropriate normalization operation, as a relative value of the sensitivity, for the displacement demodulation algorithm (31) to call; the displacement demodulation algorithm (31) will first reversely calculate the light intensity E received by each pixel in the photoelectric detection device (14) according to the voltage signal output by the photoelectric detection device (14) and the sensitivity calculated by the sensitivity demodulation algorithm (32). m , and then by substituting formulas (6) and (7) into formulas (8) and (9), the corrected light intensity E and phase shift φ are demodulated, and the displacement x on the high-precision grating scale incremental code track is obtained.

[0011] The collimated light source (11) in the system can be any one of the light source systems, such as a combination system of a light emitting diode (LED) and a special lens, or a light source system formed by collimating / expanding a laser. The collimated light source (11) is characterized in that it has a relatively small beam divergence angle, and its beam divergence angle (full width at half maximum) is less than 10°. Preferably, the collimated light source (11) should also have a relatively narrow spectral range, or be a monochromatic light source, and its spectral width (full width at half maximum) is less than 35nm. Preferably, the collimated light source (11) should also have a relatively large aperture, and the output light field can illuminate the width of the incremental code channel of the scale grating (12), and can also illuminate the incremental code channel window range of the indicator grating (13).

[0012] The scale grating (12) in the system is a transmission grating or a reflection grating that can undergo one-dimensional relative displacement with the collimated light source (11) and the photoelectric detection device (14), and can modulate the light field along with the relative displacement and project it onto the indicating grating (13); the characteristic of the scale grating (12) is that the scale grating (12) uses a metal material or a glass material as a substrate, and is plated with periodic polished metal stripes or dielectric stripes, and can periodically reflect or transmit the incident light beam; the grating pitch in the incremental code channel area on the scale grating (12) can be 20 μm, 30 μm, 40 μm, 50 μm, etc.; correspondingly, the width of a single stripe can be 10 μm, 15 μm, 20 μm, 25 μm, etc.

[0013] The indicator grating (13) in the system is a transmission grating fixed relative to the collimated light source (11) and the photoelectric detection device (14), and can filter the light field spatially modulated by the scale grating (12); the characteristic is that the indicator grating (13) uses glass as a substrate, on which periodic dielectric stripes are plated, and can transmit the incident light beam; the grating pitch in the incremental code channel area on the indicator grating (13) can be 20μm, 30μm, 40μm, 50μm, etc.; correspondingly, the width of a single stripe can be 10μm, 15μm, 20μm, 25μm, etc.; the stripes in the incremental code channel area on the indicator grating (13) must be consistent with the stripe period and width in the incremental code channel area on the scale grating (12).

[0014] The adjustable driving power supply (21) in the system is an excitation source capable of providing stable electrical stimulation for the uniform light source (22), and can continuously adjust the light intensity of the uniform light source (22); the characteristic is that the power input to the adjustable driving power supply (21) is a high-voltage industrial frequency AC power supply (220V mains) or a constant voltage DC power supply converted by a power adapter, and the output power is a constant voltage DC power supply or a constant current DC power supply, and the output voltage value or current value can be adjusted by a knob on it; preferably, the adjustable driving power supply (21) should also be a constant current source whose output voltage changes with the forward voltage drop of the uniform light source (22); preferably, the adjustable driving power supply (21) should also have a larger voltage and current adjustment range, the adjustment range is greater than 5V and 0.5A; preferably, the adjustable driving power supply (21) should also have a smaller voltage and current adjustment step, the adjustment step is less than 0.1V and 0.01A; preferably, the adjustable driving power supply (21) should also have a voltage limiting function and a current limiting function, and be protected in an open circuit and a short circuit state.

[0015] The uniform light source (22) in the system can be any one of the light source systems such as a combination system of a light emitting diode (LED) and a light guide plate or a combination system of a laser and an integrating sphere, and can provide a light field with uniform irradiance. The system is characterized in that the uniform light source (22) should have a relatively large aperture, and the light field outputted can illuminate the entire photosensitive area of ​​the photoelectric detection device (14) and the photoelectric detection device (24). Preferably, the uniform light source (22) should also have the same spectral characteristics as the collimated light source (11), that is, the central wavelengths of the uniform light source (22) and the collimated light source (11) are equal, and both have relatively narrow spectral ranges, and the spectral width (full width at half maximum) is less than 35nm.

[0016] The power meter (23) in the system is a measuring device that can characterize the optical power (or irradiance) at the position of its probe; it is characterized in that the power meter (23) should have a large range and a small resolution, its range should be larger than the adjustable range of the combined system of the adjustable driving power supply (21) and the uniform light source (22), and its resolution should be smaller than the adjustable step size of the combined system of the adjustable driving power supply (21) and the uniform light source (22); preferably, the power meter (23) should also be calibrated according to the wavelength of the light field to be measured; preferably, the power meter (23) should also have a small measurement uncertainty, which is usually less than 5%; preferably, the power meter (23) should also take repeated measurements for the same light intensity and take the average value as the optical power output.

[0017] The photoelectric detection device (14) and the photoelectric detection device (24) in the system can be any one of the photoelectric detection systems such as a photovoltaic cell (array) or a photoelectric diode (array) with a reverse bias and a transimpedance amplifier. The same photoelectric detection device is respectively applied to modules such as a grating scale module (1) and a sensitivity calibration method (2), and receives the spatially modulated light field signal passing through the scale grating (12) and the indicator grating (13), and the spatially uniform light field signal emitted by the uniform light source (22); the photoelectric detection device is characterized in that the photoelectric detection device The photoelectric detection device (14) and the photoelectric detection device (24) can generate a resolvable voltage signal according to the light power (or irradiance) incident thereon; preferably, the voltage signal generated by the photoelectric detection device (14) and the photoelectric detection device (24) should also be proportional to or linearly related to the light power (or irradiance) incident thereon; preferably, each pixel in the photoelectric detection device (14) and the photoelectric detection device (24) should also have the same photosensitive surface size; preferably, each pixel in the photoelectric detection device (14) and the photoelectric detection device (24) should also have the same dark current.

[0018] The sensitivity demodulation algorithm (32) in the system can calculate the sensitivity of each pixel of the photoelectric detection device (24) as a correction parameter of the displacement demodulation algorithm (31); it is characterized in that after receiving the light power output by the power meter (23) and the voltage signal of each pixel of the photoelectric detection device (24), the sensitivity demodulation algorithm (32) takes the light power as the independent variable and the voltage as the dependent variable, records the mapping relationship between the voltage and the light power of each pixel in the form of discrete points, and then performs linear fitting on the recorded discrete points of each pixel, extracting the slope of each pixel as the sensitivity (the unit of the slope is V / W at this time); preferably, the sensitivity demodulation algorithm (32) The algorithm (32) performs appropriate normalization operations on the extracted sensitivity and converts it into a floating point number with a unit of 1 as a relative value of the sensitivity; preferably, the sensitivity demodulation algorithm (32) should also calibrate the sensitivity of the photoelectric detection device (24) once before the system leaves the factory through a sensitivity calibration method (2) and other modules, and store it in a non-volatile memory (NVM) on the high-precision grating scale displacement demodulation algorithm (3) for use by the displacement demodulation algorithm (31); preferably, the sensitivity demodulation algorithm (32) should also calibrate and update the sensitivity once every period of time through the sensitivity calibration method (2), and the time is usually one year.

[0019] The displacement demodulation algorithm (31) in the system can calculate and output the displacement of the grating scale after sensitivity correction error according to the voltage of each pixel output by the photoelectric detection device (14) and the sensitivity of each pixel extracted by the sensitivity demodulation algorithm (32); in a specific implementation, after the displacement demodulation algorithm (31) receives the voltage of each pixel output by the photoelectric detection device (14) and the sensitivity of each pixel extracted by the sensitivity demodulation algorithm (32), it divides the voltage by the sensitivity pixel by pixel to reversely calculate the light intensity (relative value) of each pixel; the displacement demodulation algorithm (31) accumulates the light intensity of all pixels representing each phase to obtain the light intensity E of each phase. m ; The displacement demodulation algorithm (31) converts the light intensity E of each phase m Substitute into formulas (6) and (7) to calculate the cosine part I and the sine part Q of the phase shift; the displacement demodulation algorithm (31) substitutes the cosine part I and the sine part Q of the phase shift into formulas (8) and (9), calculates the light intensity E and the phase shift φ, and converts the phase shift φ into the displacement x of the grating scale incremental code channel, and finally outputs the light intensity E and the displacement x of the grating scale incremental code channel. At this time, the displacement is a high-precision displacement after sensitivity correction of nonlinear errors.

[0020] The sensitivity calibration method (2), sensitivity demodulation algorithm (32) in the system and the method of calculating light intensity E and phase shift φ using formulas (6), (7), (8) and (9) involved in the present invention can be applied not only to the displacement x demodulation of the grating scale system, but also to the distance demodulation of the indirect time of flight (iToF) ranging system based on time modulation and other phase demodulation-based algorithm systems. (IV) Description of the drawings

[0021] Figure 1 It is a system block diagram of the high-precision grating scale displacement demodulation algorithm and sensitivity calibration method. The whole system is composed of a grating ruler module (1), a sensitivity calibration method (2), and a high-precision displacement demodulation algorithm (3); wherein the grating ruler module (1) comprises a collimated light source (11), a scale grating (12), an indicator grating (13), and a photoelectric detection device (14); the photoelectric detection device (14) is a calibrated photoelectric detection device, and can output a resolvable voltage according to the relative displacement of the scale grating (12) and the indicator grating (13) thereon; wherein the sensitivity calibration method (2) comprises an adjustable driving power supply (21), a uniform light source (22), a power meter (23), and a photoelectric detection device (24); the power meter (23) can output the detected light power of the uniform light source (22); the photoelectric detection device (24) is a photoelectric detection device to be calibrated, and can output a resolvable voltage according to its sensitivity and the light power of the uniform light source (22); wherein the high-precision displacement algorithm (3) comprises a displacement demodulation algorithm (31) and a sensitivity demodulation algorithm (32), and can output light intensity and displacement after sensitivity correction.

[0022] Figure 2 It is the moiré fringes produced by the overlap of two black and white gratings with different grating pitches and grating line directions. According to the geometric shading principle, the spacing A and direction Θ of the moiré fringes satisfy the equations shown in formulas (1) and (2); the spacing of the moiré fringes is usually much larger than the grating pitch of each of the two gratings. When any of the gratings moves at the level of pitch a, the moiré fringes will shift at the level of fringes A, thus realizing the amplification of the displacement signal of the tiny grating scale incremental code channel.

[0023] Figure 3The phase shift φ (a) and its error with the true value (b) are calculated before (blue dot) and after (orange dot) sensitivity correction as the grating scale moves. In the case of optical gate moiré fringes, the sensitivities of phase 0, phase 1, phase 2, and phase 3 extracted by the sensitivity demodulation algorithm (32) are 1.05, 0.99, 1.02, and 0.95, respectively. The phase shift φ calculated by the traditional displacement demodulation algorithm (algorithm before correction), that is, the cosine part I and the sine part Q of the phase shift are calculated using formulas (13) and (14), and then substituted into formula (9) has a nonlinear error with the true value; however, the nonlinear error of the phase shift φ calculated by the high-precision displacement demodulation algorithm (algorithm after correction) of the present invention, that is, the cosine part I and the sine part Q of the phase shift are calculated using formulas (6) and (7) after the voltage signal is reversely converted into light intensity, and then substituted into formula (9) is eliminated.

[0024] Figure 4 Schematic diagram of data acquisition by the photoelectric detection device (14) when the moiré fringes are not the shutter moiré fringes (i.e., there is a spatially modulated light field). (a) represents the spatially modulated light field that changes with the displacement of the grating scale incremental code track when the moiré fringes are present. The phase shift φ of the moiré fringes and the displacement x of the grating scale incremental code track satisfy: φ = 2πx / a. (b) represents the arrangement of each pixel in the photoelectric detection device (14) in the direction perpendicular to the moiré fringes. The pixel width is 1 / 4 of the fringe spacing. Each phase is represented by multiple pixels, where the center of the pixel width representing phase 0 should be aligned with the peak position of the moiré fringes when x = 0. (c) is the light power condition irradiating each pixel of the photoelectric detection device (14). Each pixel will output a resolvable voltage signal according to the light power irradiated on its photosensitive area. (V) Specific implementation methods

[0025] Embodiment 1:

[0026] This embodiment is a multi-field scanning embodiment, that is, the moiré fringes are light gate moiré fringes, and the fringe spacing tends to infinity, and is an embodiment for extracting the incremental code track displacement of the grating scale. The high-precision grating scale displacement demodulation algorithm and sensitivity calibration method are composed of a grating scale module (1), a sensitivity calibration method (2), and a high-precision displacement demodulation algorithm (3). The indicator grating (13) in the grating scale module (1) includes 4 groups of transmission gratings, which represent phase 0, phase 1, phase 2, and phase 3 in turn. The grating pitch and grating line direction of each grating are consistent with those in the scale grating (12), and the grating lines of the gratings representing adjacent phases differ in turn by 1 / 4 of the grating pitch. The 4 groups of gratings of the indicator grating (13) are followed by 4 pixels of the photoelectric detection device (14), which represent phase 0, phase 1, phase 2, and phase 3 in turn. Each pixel is aligned with the grating representing each phase on the indicator grating (13). The sensitivities (relative values) of the pixels representing phase 0, phase 1, phase 2, and phase 3 respectively extracted by the sensitivity demodulation algorithm (32) are 1.05, 0.99, 1.02, and 0.95 respectively.

[0027] Figure 3 (a) The blue dots show the voltage signals U0, U1, U2, and U3 received from the photoelectric detection device (14) representing each phase pixel as the incremental code track of the grating scale moves. The cosine part I and the sine part Q of the phase shift calculated by the traditional algorithm, that is, formulas (13) and (14), are substituted into formula (9) to calculate the phase shift φ. Figure 3 (b) The blue dot represents the error between the phase shift φ and the true value. Figure 3 It is not difficult to see in (b) that, affected by the difference in pixel sensitivity of the photoelectric detection device (14) representing each phase, the phase shift error calculated using the traditional grating scale displacement demodulation algorithm is at most greater than 0.08. Converted to the grating scale displacement x, there is a nonlinear error of at most greater than 1% of the grating pitch.

[0028] Figure 3 (a) The orange dots show the voltage signals U0, U1, U2, U3 received from the photoelectric detection device (14) representing each phase pixel as the incremental code track of the grating scale is displaced. The high-precision displacement demodulation algorithm (3) of the present invention is used, that is, firstly, the voltages U0, U1, U2, U3 of each pixel are reversely converted into the light intensities E0, E1, E2, E3 of each pixel, and then the cosine part I and the sine part Q of the phase shift calculated by formulas (6) and (7) are substituted into the phase shift φ calculated by formula (9). Figure 3 (b) The orange dot represents the error between the phase shift φ and the true value. Figure 3 From the blue and orange dots in (b), it is easy to see that by using the high-precision displacement demodulation algorithm (3) of the present invention, the nonlinear error caused by the difference in pixel sensitivity of the photoelectric detection device (14) representing each phase is eliminated.

[0029] Embodiment 2:

[0030] This embodiment is a single field scan, that is, there is Figure 4 (a) shows a spatially modulated light field that changes with the displacement of the grating scale incremental code track, and an embodiment of extracting the displacement of the grating scale incremental code track. The high-precision grating scale displacement demodulation algorithm and sensitivity calibration method are composed of a grating scale module (1), a sensitivity calibration method (2), and a high-precision displacement demodulation algorithm (3). The indicator grating (13) is a group of transmission gratings. The light field emitted by the collimated light source (11) is modulated by the scale grating (12) and the indicator grating (13) to form a light field at the photoelectric detection device (14). Figure 4 (a) shows a spatially modulated light field that changes with the displacement of the grating scale incremental code track. The photoelectric detection device (14) is a group of linear array photoelectric detection devices, such as Figure 4 As shown in (b), the pixels are arranged in sequence along the vertical direction of the moiré fringe, representing phase 0, phase 1, phase 2, phase 3, phase 0, phase 1, phase 2, phase 3, ..., and the width of each pixel is 1 / 4 of the fringe spacing. The center of the pixel width representing phase 0 should be aligned with the peak position of the moiré fringe when the incremental code displacement of the grating scale is zero. Figure 4 As shown in (c), each pixel of the photoelectric detection device (14) converts the light power irradiated in the photosensitive area thereof into a voltage signal and outputs it to the displacement demodulation algorithm (31).

[0031] After receiving the voltage signals of each pixel of the photoelectric detection device (14), the displacement demodulation algorithm (31) reversely calculates the light intensity of each pixel according to the sensitivity of each pixel extracted by the sensitivity demodulation algorithm (32). Then, E0, E1, E2, and E3 are calculated respectively, where E0 is the sum of the light intensities of all pixels representing phase 0, E1 is the sum of the light intensities of all pixels representing phase 1, and so on. After obtaining the values ​​of E0, E1, E2, and E3 respectively, the cosine part I and the sine part Q of the phase shift are calculated using formulas (6) and (7), and are substituted into formula (9) to calculate the phase shift φ. Compared with the first embodiment, the phase shift is the phase shift calculated by formulas (6), (7), and (9), and is the phase shift that eliminates the nonlinear error caused by the difference in sensitivity of each pixel of the photoelectric detection device (14).

[0032] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A high-precision grating scale displacement demodulation algorithm and sensitivity calibration method, characterized in that: The system is composed of a grating ruler module (1), a sensitivity calibration method (2), and a high-precision displacement demodulation algorithm (3). The grating ruler module (1) comprises a collimated light source (11), a scale grating (12), an indicating grating (13), and a photoelectric detection device (14); the sensitivity calibration method (2) comprises an adjustable driving power supply (21), a uniform light source (22), a power meter (23), and a photoelectric detection device (24); the high-precision displacement algorithm (3) comprises a displacement demodulation algorithm (31) and a sensitivity demodulation algorithm (32); wherein the photoelectric detection device (14) and the photoelectric detection device (24) are the same photoelectric detection device when they are respectively applied to the grating ruler module (1) and the sensitivity calibration method (2), and are respectively used to receive a spatially modulated light field and a spatially uniform light field. field, both of which can output a resolvable voltage signal according to the irradiance of the light field incident thereon; the sensitivity demodulation algorithm (32) in the high-precision displacement demodulation algorithm (3) can calculate the sensitivity (usually in V / W) of each pixel of the photoelectric detection device according to the voltage signal output by the photoelectric detection device (24) and the light power output by the power meter (23); the calculated sensitivity can be converted into a floating point number with a unit of 1 through appropriate normalization operations as a relative value of the sensitivity for use by the displacement demodulation algorithm (31); the displacement demodulation algorithm (31) will first reversely calculate the light intensity E received by each pixel in the photoelectric detection device (14) according to the voltage signal output by the photoelectric detection device (14) and the sensitivity calculated by the sensitivity demodulation algorithm (32) m , calculate the cosine part I and the sine part Q by the formula: Substitute it into the formula and demodulate the corrected light intensity E and phase shift φ: Then extract the displacement x on the high-precision grating scale incremental code track: 。 2. The high-precision grating scale displacement demodulation algorithm and sensitivity calibration method according to claim 1, characterized in that: The collimated light source (11) in the system can be any one of the light source systems, such as a light source system formed by a light emitting diode (LED) and a special lens or a light source system formed by a laser after collimation / expansion; wherein: the collimated light source (11) has a relatively small beam divergence angle, and its beam divergence angle (full width at half maximum) is less than 10°; preferably, the collimated light source (11) should also have a relatively narrow spectral range, or be a monochromatic light source, and its spectral width (full width at half maximum) is less than 35nm; preferably, the collimated light source (11) should also have a relatively large aperture, and the output light field can illuminate the width of the incremental code channel of the scale grating (12), and can also illuminate the incremental code channel window range of the indicator grating (13).

3. The high-precision grating scale displacement demodulation algorithm and sensitivity calibration method according to claim 1, characterized in that: The scale grating (12) in the system is a transmission grating or a reflection grating that can undergo one-dimensional relative displacement with the collimated light source (11) and the photoelectric detection device (14), and can modulate the light field along with the relative displacement and project it onto the indicating grating (13); wherein: the scale grating (12) uses a metal material or a glass material as a substrate, and is plated with periodic polished metal stripes or dielectric stripes, and can periodically reflect or transmit an incident light beam; the grating pitch in the incremental code channel area on the scale grating (12) can be 20 μm, 30 μm, 40 μm, 50 μm, etc.; correspondingly, the width of a single stripe can be 10 μm, 15 μm, 20 μm, 25 μm, etc.

4. The high-precision grating scale displacement demodulation algorithm and sensitivity calibration method according to claim 1, characterized in that: The indicator grating (13) in the system is a transmission grating fixed relative to the collimated light source (11) and the photoelectric detection device (14), and can filter the light field spatially modulated by the scale grating (12); wherein: the indicator grating (13) uses glass as a substrate, on which periodic dielectric stripes are plated, and can transmit the incident light beam; the grating pitch in the incremental code channel area on the indicator grating (13) can be 20 μm, 30 μm, 40 μm, 50 μm, etc.; correspondingly, the width of a single stripe can be 10 μm, 15 μm, 20 μm, 25 μm, etc.; the stripes in the incremental code channel area on the indicator grating (13) must be consistent with the stripes in the incremental code channel area on the scale grating (12) The period and width of the stripes are consistent.

5. The high-precision grating scale displacement demodulation algorithm and sensitivity calibration method according to claim 1, characterized in that: The adjustable driving power supply (21) in the system is an excitation source capable of providing stable electrical stimulation for the uniform light source (22), and can continuously adjust the light intensity of the uniform light source (22); wherein: the power input to the adjustable driving power supply (21) is a high-voltage industrial frequency AC power supply (220V mains) or a constant voltage DC power supply converted by a power adapter, and the output power is a constant voltage DC power supply or a constant current DC power supply, and the output voltage value or current value can be adjusted by a knob on it; preferably, the adjustable driving power supply (21) should also be a constant current source whose output voltage changes with the forward voltage drop of the uniform light source (22); preferably, the adjustable driving power supply (21) should also have a larger voltage and current adjustment range, the adjustment range is greater than 5V and 0.5A; preferably, the adjustable driving power supply (21) should also have a smaller voltage and current adjustment step, the adjustment step is less than 0.1V and 0.01A; preferably, the adjustable driving power supply (21) should also have a voltage limiting function and a current limiting function, and be protected in an open circuit and a short circuit state.

6. The high-precision grating scale displacement demodulation algorithm and sensitivity calibration method according to claim 1, characterized in that: The uniform light source (22) in the system can be any one of the light source systems such as a combination system of a light emitting diode (LED) and a light guide plate or a combination system of a laser and an integrating sphere, and can provide a light field with uniform irradiance; wherein: the uniform light source (22) should have a larger aperture, and the light field outputted can illuminate the entire photosensitive area of ​​the photoelectric detection device (14) and the photoelectric detection device (24); preferably, the uniform light source (22) should also have the same spectral characteristics as the collimated light source (11), that is, the central wavelengths of the uniform light source (22) and the collimated light source (11) are equal, and both have a narrow spectral range, and the spectral width (full width at half maximum) is less than 35nm.

7. The high-precision grating scale displacement demodulation algorithm and sensitivity calibration method according to claim 1, characterized in that: The power meter (23) in the system is a measuring device that can characterize the optical power (or irradiance) at the position of its probe; wherein: the power meter (23) should have a large range and a small resolution, and its range should be larger than the adjustable range of the combined system of the adjustable driving power supply (21) and the uniform light source (22), and its resolution should be smaller than the adjustable step length of the combined system of the adjustable driving power supply (21) and the uniform light source (22); preferably, the power meter (23) should also be calibrated according to the wavelength of the light field to be measured; preferably, the power meter (23) should also have a small measurement uncertainty, which is usually less than 5%; preferably, the power meter (23) should also take repeated measurements for the same light intensity and take the average value as the optical power output.

8. The high-precision grating scale displacement demodulation algorithm and sensitivity calibration method according to claim 1, characterized in that: The photoelectric detection device (14) and the photoelectric detection device (24) in the system can be any one of the photoelectric detection systems such as a photovoltaic cell (array) or a photodiode (array) with a reverse bias and a transimpedance amplifier. The same photoelectric detection device is respectively applied to modules such as a grating scale module (1) and a sensitivity calibration method (2), and receives the spatially modulated light field signal passing through the scale grating (12) and the indicator grating (13), and the spatially uniform light field signal emitted by the uniform light source (22); wherein: the photoelectric detection device (14) and the photoelectric detection device (24) can generate a resolvable voltage signal according to the light power (or irradiance) incident thereon; preferably, the voltage signal generated by the photoelectric detection device (14) and the photoelectric detection device (24) should also be proportional to or linearly related to the light power (or irradiance) incident thereon; preferably, each pixel in the photoelectric detection device (14) and the photoelectric detection device (24) should also have the same photosensitive surface size; preferably, each pixel in the photoelectric detection device (14) and the photoelectric detection device (24) should also have the same dark current.

9. The high-precision grating scale displacement demodulation algorithm and sensitivity calibration method according to claim 1, characterized in that: The sensitivity demodulation algorithm (32) in the system can calculate the sensitivity of each pixel of the photoelectric detection device (24) as a correction parameter of the displacement demodulation algorithm (31); wherein: after receiving the light power output by the power meter (23) and the voltage signal of each pixel of the photoelectric detection device (24), the sensitivity demodulation algorithm (32) takes the light power as the independent variable and the voltage as the dependent variable, and records the mapping relationship between the voltage and the light power of each pixel in the form of discrete points, then performs linear fitting on the recorded discrete points of each pixel, and extracts the slope of each pixel as the sensitivity (the unit of the slope is V / W at this time); preferably, the sensitivity demodulation algorithm Method (32) performs appropriate normalization operations on the extracted sensitivity and converts it into a floating point number with a unit of 1 as a relative value of the sensitivity; preferably, the sensitivity demodulation algorithm (32) should also calibrate the sensitivity of the photoelectric detection device (24) once before the system leaves the factory through a sensitivity calibration method (2) and other modules, and store it in a non-volatile memory (NVM) on the high-precision grating scale displacement demodulation algorithm (3) for use by the displacement demodulation algorithm (31); preferably, the sensitivity demodulation algorithm (32) should also calibrate and update the sensitivity once every period of time through the sensitivity calibration method (2), and the time is usually one year.

10. The high-precision grating scale displacement demodulation algorithm and sensitivity calibration method according to claim 1, characterized in that: The displacement demodulation algorithm (31) in the system can calculate and output the displacement of the grating scale after sensitivity correction error according to the voltage of each pixel output by the photoelectric detection device (14) and the sensitivity of each pixel extracted by the sensitivity demodulation algorithm (32). In specific implementation, after the displacement demodulation algorithm (31) receives the voltage of each pixel output by the photoelectric detection device (14) and the sensitivity of each pixel extracted by the sensitivity demodulation algorithm (32), it divides the voltage by the sensitivity pixel by pixel to reversely calculate the light intensity (relative value) of each pixel. The displacement demodulation algorithm (31) accumulates the light intensity of all pixels representing each phase to obtain the light intensity E of each phase. m ; The displacement demodulation algorithm (31) is based on the light intensity E of each phase m , calculate the cosine part I and the sine part Q of the phase shift; the displacement demodulation algorithm (31) calculates the light intensity E and the phase shift φ according to the cosine part I and the sine part Q of the phase shift, and converts the phase shift φ into the displacement x of the grating scale incremental code channel, and finally outputs the light intensity E and the displacement x of the grating scale incremental code channel. At this time, the displacement is a high-precision displacement after sensitivity correction of nonlinear errors.

11. A high-precision grating scale displacement demodulation algorithm and sensitivity calibration method according to claims 1-10, characterized in that: The sensitivity calibration method (2), sensitivity demodulation algorithm (32) in the system and the method for calculating the cosine part I of the phase shift, the sine part Q of the phase shift, the light intensity E and the phase shift φ involved in the present invention can be applied not only to the displacement x demodulation of the grating scale system, but also to the distance demodulation of the indirect time of flight (iToF) ranging system based on time modulation and other algorithm systems based on phase demodulation.

Images